How machine learning can improve COVID testing
- By Susan Miller
- Jul 22, 2020
On June 18, the Food and Drug Administration authorized the use of pooled testing for identifying COVID-19 infections. The method allows up to four swabs to be tested at once – a strategy that is expected to greatly expand frequent testing to larger sections of the population.
The idea is that if a bundled sample comes back positive, then all the individuals in that sample will need to be tested separately. If a bundled sample comes back clean, however, that’s four people who don’t need to be tested further, saving public health officials time and money.
The FDA said it expects pooling will allow virus identification with fewer tests, which means more tests could be run at once, fewer testing supplies would be consumed and patients could likely receive results more quickly. The pooling strategy will be most efficient in areas where the outbreak is under control, meaning where only a small percentage of test subjects are expected to be infected, the FDA acknowledged.
Researchers from the University of California, Berkeley, think even more people can be tested by combining pooling with machine learning tools that estimate the risk of COVID for each person to be tested, according to an article they authored in MIT Technology Review.
“Take, for example, data on the home zip code of an individual,” the researchers said in their research paper. “While zip code alone is certainly informative, interactions with other observable factors such as age, travel or consumption patterns are likely to improve model performance in terms of prediction for day-specific risk (e.g. living in a location with high prevalence increases ones risk but more so if a person frequently eats out and socializes and even more so if they did so in the last two days).”
“Using publicly available data from employers and schools, epidemiological data on local infection and testing rates, and more sophisticated data on travel patterns, social contacts, or sewage, if available, modelers can predict anyone's risk of having COVID-19 on a day-by-day basis,” the researchers wrote in MIT Technology Review.
With that information in hand, public health officials could test pools of optimal individuals, winnowing out likely carriers of the virus based on location, demographics, age, job type, living situation, along with past infection data. Conducting pooled testing on those with low risk of infection will therefore greatly expand the speed and efficiency of testing.
In fact, high-frequency pooled testing augmented with machine learning will actually cost less, the researchers said. “Large populations can be tested weekly or even daily, for as low as $3 to $5 per person per day,” they wrote. Plus, they said their analysis showed “testing daily costs only twice as much as testing monthly. And daily testing can actively suppress the virus, whereas monthly testing really only allows us to see how badly things have gone.”
Susan Miller is executive editor at GCN.
Over a career spent in tech media, Miller has worked in editorial, print production and online, starting on the copy desk at IDG’s ComputerWorld, moving to print production for Federal Computer Week and later helping launch websites and email newsletter delivery for FCW. After a turn at Virginia’s Center for Innovative Technology, where she worked to promote technology-based economic development, she rejoined what was to become 1105 Media in 2004, eventually managing content and production for all the company's government-focused websites. Miller shifted back to editorial in 2012, when she began working with GCN.
Miller has a BA and MA from West Chester University and did Ph.D. work in English at the University of Delaware.
Connect with Susan at [email protected] or @sjaymiller.